Wikimedia CH/Offline Dissemination

The Offline Dissemination Program (ODP) wants to increase the audience of free content (in particular Wikipedia) using offline technologies (Kiwix).

The program was launched by WikimediaCH in 2014 and confirmed for 2015.

If you would like to get active in the Offline Dissemination Program (ODP) of WMCH, please get in touch with Muriel.

If you have technical questions about KIWIX, you can contact Emmanuel.

2015

Plan

WMCH has launched the Offline Dissemination Program (ODP) in 2014 with the aim to make Wikipedia accessible (with KIWIX software) in areas with no internet connection or where the internet is being cenzored or filtered. This program has been strongly welcomed by both the community and stakeholders outside the movement. Offline solutions are not only a way to improve our outreach, but also to provide a possibiltiy to acquire new members to our movement: the moment people start using Wikipeida offline, they see it's added value and want to contribute themselves.

Due to the ("offline") nature of technology, gathering usage statistics is complicated & challenging. But, even with only a few metrics available, we are able to proove the past success of the program:

Nearly 50K Android App Downloads
Country-wide deployment of Wikipedia offline in swiss prisons
Further numbers (TODO to complete)

Based on these facts, Wikimedia CH wants to continue investing in offline dissemination of Wikipedia in 2015 and make its development a high priority of the associations overall activities.

It's important to understand that the added value of the ODP is to provide technical offline solutions to third parties. Therefore, it's not only important to have the best technology available, but also to develop strong partnerships with organizations operating on the field. In 2015, the ODP will consequently be based on the following 3 pilars: Projects collaboration, R&D and Communication & External Founding.

Projects collaboration

Here are an overview of the planned activities and deliverables.

Develop partnerships with NGOs which are on place, to deploy Wikipedia offline

Having a network of NGOs which are there on place near to the population suffering of lack of connectivity is essential. Therefore, we need to increase our number of "partner big deployments". Here are a short list of potential projects we are currently in talk with:

Wikipedia zero: The successful WMF project currently evaluates the opportunity to make deal with operators/vendors to pre-ship the Wikipedia app with content (for example Wikipedia medicine articles in Africa).
Malebooks: This project plans to rollout 3000 e-ink readers in Mali. We have ported Kiwix to an e-ink device and the pilot is currently ongoing.
Afripédia: This project has deployed Wikipedia offline in more than 50 African universities for the last two years. Discussions have started to design a new version of the project, one of the key point is to find solution to allow any organisation to easily get the tech and deploy it.
Outernet: Outernet plans to provide Wikipedia offline via satellite, we are currently evaluating with them the best technical solution.

In addition, we want to support 10 grass-root partnership-initiatives of volunteers. That means that if wikimedians come with small deployments initiatives, we want to be able to support them too.

Wikipedia in Jail

Wikipedia in Jail is a success in Switzerland, we want to scale it:

In Switzerland: Accompany and further develop the partnership
Worldwide: Try to scale best-practice of "Wikipedia offline in Jail in Switzerland" to at least 2 other countries.

R&D

The ODP program is mainly based on softwares (Kiwix and openZIM) which are developed and maintained by volunteers. Although we are pretty satisfied by this volunteer run projects, there are important things which are difficult (and partly impossible) to get done correctly:

"Boring task" (sysadmin work, testing, release management, ...), due to the lack of motivated volunteers (they prefer to work on challenging technical problems).
Reliable support, because volunteers are not always available during office hours and cannot be contracted.
Internships, because there is nobody to recruit and mentor them.
Customer oriented developments/polishing, because the interest of the market is not always the interest of the code hackers.

Due to these weaknesses, professional offers or big partnerships are impossible, even though the demand is there. These big deals/deployments are essential to push the software development. Worse, without them it's really difficult to reach big amounts of people and simply impossible to measure our impact on site. That's why the ODP wants to recruit two people in 2015: a developer (100%) and a R&D manager (20%).

This small team will have the following main technical goals for 2015:

Get Wikimedia project ZIM files generated in the Wikimedia datacenter. This is essential to be able to release quickly, often and without any hand work new ZIM files for all our projects. By doing this we will stop one of the most frequently criticised aspects with regard to the Kiwix project.
Introduce automated tests in the project. Kiwix supports 5 different operating systems and with the probable introduction of iOS in 2015 the project is going to be unmaintainable without a rigorous test automation. One of the first tasks will be to set up a test strategy and start to implement it with Kiwix for Android.

Improving Kiwix for Android is a tech. priority for 2015
Having improved the Kiwix Android app. in 2014, more smartphones than personal computers were sold. It's essential that we start to fill the feature and quality gap between the Android app and the Kiwix desktop version. To introduce a download manager and a fulltext search is a priority.
Improve ZIM updatability: the ZIM files which store the offline wikipedia articles are not easy to update at the moment. This is a repeated feature request from customers in order to be able to do incremental updates. This is, for example, a requirement of the "Wikipedia Zero" team which thinks about shipping a content package with the Wikipedia app.
Improve the ZIM file creation toolset. The ZIM format is recognized to be an efficient solution. Although many people are interested to use the toolset, its ways of usage are quite limited. To push the publication of new ZIM files (non Wikimedia content) and encourage new developers to join in, we need to improve it. The top priority is to help in the development of the OCG and add the support for ZIM, that probably means the creation of the nodezim binding. The second priority is to create phpzim, a binding which is mandatory to create new export plugin for CMS (like Wordpress for example).

But, having a professional's taskforce on Kiwix does not mean that the project abandons his volunteer based development model. Replacing volunteer workforce by professional workforces is simply not affordable and against the philosophy of both the movement and WMCH. That's why we will still continue to focus on volunteers and will:

organize at least two offline hackathons during the year (focus will be given on package additional free content in the ZIM format),
sponsor volunteer venues to Wikimania & Wikimedia hackathons,
sponsor hardware to push Kiwix development on mobile devices,
push the tech. communication and documentation in direction of volunteers.

Communication & External Funding

The main driver behind our "Communication & External Funding" move is to ensure that our ODP program is fully independent of FDC funding. Though these goals are set long term, we believe that today's actions will be the very first milestone to a sustainable financial model.

Professional offer

In order to achieve this goal, we will provide a "professional offer" package:

Offering an online, easy-to-use web-shop to order the Kiwix offerings as a whole fro the autumn 2015
Create sub-offering such as consulting services (as-is state analysis, gap analysis, technical support, general consulting regarding the product and design)
Create SLA-based product support (On-site/Off-site delivery and maintenance support)
Ensure an dedicated marketing and sales communication channel with existing and prospective customers, while dynamically steering our sales strategy

Our KPI: Total Sales Rev +20k

Institutional donors

We will also dedicate part of our energy to raise money from "institutional/big donors":

Develop a big donators strategy
Create communication material dedicated to institution
Take contact with at least 20 institution and advert the ODP and build a donators network

Our KPI: Find external funding of at least 20K$ from foundations and/or other institutions

Other

To support this effort we will attend at least two non-tech and non-wikimedians events.

FDC Annual Plan Grant proposal

Develop a sustainable approach to distribute wikipedia to those who have no access

In 2015 WMCH will improve the efficiency of the communication around Wikipedia Offline by distributing to key stakeholders the online and printed guides created in 2014
We will continue to promote the offline solution to a broader community by attending at least 2 non wikimedia/non tech events
we will continue our effort to increase the content offer by adding at least two collections out of Wikimedia projects (In 2014 the Gutenberg collection and the TED talks have been added)
we will continue our partnership strategy by :

Set-up partnerships with at least two new other Wikimedia Organizations

including the replication of the wikipedia in Jails program

engage in at least one new partnership with a NGO
engage in at least one new partnership with institutional actors from Francophonie
Improve the link with the Wikipedia Zero program

We will improve the offline solution by :

continuing our current effort within the tech & Wikimedia communities by promoting the participation of volunteers to tech events like the Wikimedia Hackathon, Linuxtag or the Wikimania Hacakthon.
organizing two specific hackathons dedicated to the Offline technology
recruit a developer to take in charge the tasks that enable volunteers to make a better work

by the end of 2015 the Wikimedia project ZIM files will be generated in the Wikimedia datacenter.
in order to improve the maintenance of Kiwix, in 2015 we will set up a Kiwix automated test strategy and conduct a pilot experiment with Kiwix for Android before the end of the year, before deployed it to the 3 (may be 4 since then) other different operating systems.

Improving Kiwix for Android is a tech. priority for 2015
we will improved the Kiwix Android app by the introduction of a content download manager and a fulltext search functionality.
we will introduce by the end of 2015 the ZIM incremental updatability.
To push the publication of new ZIM files (non Wikimedia content) and encourage new developers to join in, we will improve the ZIM file creation toolset by helping the development of the OCG and add the support for ZIM.

2016

Perspectives

End of 2016, the ODP program should be one of the pillar of WMCH activity:

have a working R&D department and office (~3 FT)
be self financed for at least 30% of its budget
have a workable professional offer (staff/portfolio/communication material)
have a clear network of partners/customers deploying our solutions